In this revised application, we propose to conduct annual assessments through 18 years of age to complete an ongoing longitudinal study of the predictive validity of DSM-IV criteria for attention-deficit/ hyperactivity disorder (ADHD) when it is first diagnosed at 4-6 years of age. It is important to evaluate the predictive validity of ADHD in young children, partly because of the documented increase in the use of psychoactive medications in preschool children. Because all treatments confer some risk, it is essential to know if the DSM-IV criteria for ADHD accurately identify younger children with significant functional impairment and adverse long-term outcomes. The sample consists of 125 ADHD probands and 130 well-matched nonreferred comparison children who are currently being assessed in their tenth annual assessments. In order to inform decision making for DSM-V, we will use the proposed adolescent follow-ups to address the predictive validity of the diagnostic criteria for ADHD, including symptoms and thresholds, the validity of the subtypes of ADHD, and the appropriateness of age-specific symptoms and diagnostic thresholds. The present study has the advantages of using DSM-IV diagnostic criteria and beginning annual structured diagnostic assessments at an early age. When completed, this study will provide the first detailed mapping of the developmental course of DSM-IV ADHD, including developmental progressions from ADHD to other mental health outcomes (conduct problems, unipolar depression, bipolar disorder, anxiety disorders, and substance use) from early childhood through adolescence. Because we expect ADHD probands to exhibit declining average levels of ADHD symptoms over time, we will adopt a developmental approach to predictive validity. That is, we predict that probands whose ADHD symptoms persist into adolescence will be more likely to exhibit impaired functioning and experience other adverse mental health outcomes in adolescence. If this hypothesis is confirmed, as preliminary findings suggest, we will test key hypotheses regarding early child characteristics and potential psychosocial factors that may predict future persistence versus recovery in ADHD symptoms and impairment. Similarly, we will determine if these early child characteristic and psychosocial factors allow differential prediction of other adolescent mental health outcomes. Answering these questions will inform taxonomy, etiologic theory, and assist future efforts at targeted prevention.